I would like to ask you: How? Microsoft managed to support Unicode in Notepad, starting in Windows 2000; the heuristic used when detecting character sets is fragile and has potential to break what used to work (mostly opening readmes anyway). How would support for Unix style line ending break legacy software?